) Inc. in Santa Clara, Calif., put the sequence on a computer chip. By slotting the chip into a specially outfitted computer, a scientist can identify -- from a single sample of a patient's DNA -- which of the 30,000 possible mutations of the virus that SARS victim has. This will help scientists track how the virus changes over time, identify which strains pose the most danger, and, ultimately, uncover clues to developing vaccines and cures. Roche Diagnostics will use the chip to develop a SARS test it will release this summer.
Such advances aside, the pace of IT development is lagging that of genomic discovery. Roughly 1,500 disease-causing genes have been pinpointed to date. But information about them is scattered in databases all over the world -- in libraries, government research labs, and at the companies that are making the discoveries. At present, there is no common technology platform for obtaining or sharing genomics information, so biotechs are scrambling to develop their own systems for sifting through the data. This scientific Babel impedes the flow of usable knowledge.
On the one hand, this is all a huge business opportunity. IDC estimates that the market for hardware, software, and services in the biotech industry will more than double, to $34.5 billion, by 2007. Making bioinformatics work is a huge challenge, however, because there is no formal organization or institution that is mediating among all the competing players, ranging from Sun Microsystems (SUNW
) IBM (IBM
), and Hewlett-Packard (HPQ
) to biotech startups such as Lion Bioscience (LEON
) and Rosetta Inpharmatics Inc., a unit of Merck (MRK
) & Co.
Yet there is a shared sense of mission. "People must have the capability to build complex computational models and test them against the knowledge we have of living systems," says Caroline A. Kovac, general manager for IBM Life Sciences. IBM launched the unit in 2000 to develop information-technology tools for the industry. Ultimately, Kovac imagines that biotech will have its own subset of the Internet -- a massive virtual laboratory where new drug candidates can be plugged into computerized models of diseases to see how they will perform.